DIRECCIÓN GENERAL DEL SERVICIO DE ESTUDIOS
LA COMPETITIVIDAD DE LA ECONOMÍA ESPAÑOLA
EN UN ENTORNO DE DESAPALANCAMIENTO Y
AUSTERIDAD
José Luis Malo de Molina
Director General
JORNADA DE DEBATE (BME/FEF)
Palacio de la Bolsa de Madrid
10 de octubre de 2012
MEASURING FINANCIAL COMPETENCES IN A LARGE SURVEY:
The Spanish Survey of Financial Competences
Olympia Bover (BdE), Laura Hospido (BdE, IZA), Ernesto Villanueva (BdE)
Financial Information Forum of Latin American and Caribbean Central Banks, III Meeting
Santiago, 4 - 5 October 2017
RESEARCH AND STATISTICS
OUTLINE OF THE PRESENTATION
1. Introduction
Aim of the Spanish Survey of Financial Competences
Measuring Financial Competences
Basic features
2. Contents of the Survey
Sections of the questionnaire
Specific features
3. Fieldwork
Interviewer training
Efforts to mitigate non-response
Control
Validation
4. Conclusions
2
1. INTRODUCTION: Aim of the survey
Survey on the financial competences of the population 18-79 years of age
Face-to-face survey to a large sample of individuals, questionnaire presented as CAPI
Information about the person chosen at random, but also of financial competences from a
household member informed about the household finances
Adapts a questionnaire coordinated by OECD, implemented in several countries
Initiative of the INFE (International Network of Financial Education)
First pilot in 2010, 14 countries participated
Questionnaire revised and the second measurement involves 30 countries
The core of the Spanish questionnaire is broadly based on the one provided by INFE but with
special features
General motivation: The strong correlation between the lack of basic financial
competences and poor financial outcomes
Lusardi and Mitchell (2008, 11), Alessie, Lusardi and Van Rooij (2011), Stango and Zinman (2009)
Aim: measuring financial competences, quality of decision-making and financial outcomes among the adult population in Spain
3
1. INTRODUCTION: Measuring Financial Competences
SIMPLE INTEREST RATE: QK4_a) Suppose you put €100 into a savings account with a fixed interest
rate of 2% per year. There are no fees and taxes subject to this account. You don’t make any further
payments into this account and you don’t withdraw any money. How much money would be in the
account at the end of the first year, once the interest payment is made?
Record response numerically - - -
Don't know Refuse to answer
COMPOUND INTEREST RATE: QK4_b) Once more, if you don’t make any further payments neither
withdraw any money, once the interest payment is made, how much money would be in the account
at the end of five years [remembering there are no taxes and fees and the payment is made every
year]? Would it be:
a) More than €110
b) Exactly €110
c) Less than €110
d) Or is it impossible to tell from the information given
Don't know Refuse to answer
RISK DIVERSIFICATION: QK5) I would like to know whether you think the following statement is true
or false: It is usually possible to reduce the risk of investing in the stock market by buying a wide
range of stocks and shares.
a) True
b) False
Don't know Refuse to answer
4
• Representative sample of the population of Spanish
individuals aged 18-79 living in private householdsIt excludes institutionalized population.
• Representative of each of the 17 regions as, to a large extent,
they run the educational systemSample of about 21,000 individuals, drawn from the most recent Census
Probability sampling, stratified by region and municipality size
• Reference period:Conducted between end-September 2016 and end-June 2017
Mainly reflects situation in the moment of the interview:• Some questions refer to last 12 months (whether expenses higher than
income) and up to the last 2 years (acquisition of financial assets).
• Others to next 12 months (expectations about the value of the main
residence in a year’s time).
• Some questions refer to previous two years (if acquired some financial
product).
1. INTRODUCTION: Basic features
5
2. CONTENTS OF THE SURVEY: Sections
A. Basic demographics and labor force status
Age, household composition
Employment status, occupation, probability of job loss and of finding job
B. Respondent’s portfolio and sources of information
Knows financial products, holds them, acquired them during the last 2 years?
• Checking and savings accounts, mortgages, pension and mutual funds, stocks,
bonds, consumer loans, credit cards, life and health insurance.
Information on acquisition of financial products
C. Sources of income in old age
Current or expected
D. Attitudes toward saving
E. Financial competences (H. If the informed member is not the sample member)
F. Competences in Economics
I. Main residence
• Tenure, reasons renting/purchasing, subjective expectations about growth of own
house price next two years
• If indebted: Mortgage monthly installment and original loan-to-value.
J. Expenses and financial fragility
• Qualitative: whether or not expenses exceed revenues over last 12 months and, if
yes, how financed the difference between expenses and revenues.
• Quantitative: expenses in food, education and income (the latter in ranges).
• Entered into loan arrears last year?
6
2. CONTENTS OF THE SURVEY: Special features
• Survey to individuals, but also aim to collect information about the financialcompetences of an informed household member
Ask respondent if he or she is familiarized with household’s asset and debtholdings, as well as about the reasons for financial choices
If sample individual is not, interview a person who is informed about the householdportfolio
•Household financial competences
• Complement INFE questionnaire with information on general competences
Literacy: reading and understanding a (non-financial) text
Understanding graphs:
• Non-financial: based on PIAAC assessment of numeracy and literacy(evolution of number of births)
• Financial: based on UK Financial Capability Survey (evolution of the valueof mutual funds)
• Module on housing.
Housing tenure and reasons that led to that choice
Subjective expectations about house price growth over the next year
If financed house with mortgage: loan to value at purchase and current monthlyinstalment
7
3. FIELDWORK: Training of interviewers
• Two-day training on survey content, protocol, material (cardboards)
Centralized (118 interviewers)
In mid-September, immediately before the start of fieldwork
All interviewers tested at the end of the two-day training
• Special training points in a Survey of Financial Competences:
1. Reading questions exactly as posed in questionnaire
2. Spotting basic inconsistencies and introducing comments
Always important, more so in a survey measuring financial inclusion
E.g. individual does not have a bank account, but reports holding a mortgage
Errors are common among the least financially literate respondents.
3. Imposing discipline in section of financial competences
Importance of other household members not helping in competence questions
8
3. FIELDWORK: Efforts to mitigate non-response
• Efforts to mitigate non-response:
Letter signed by Governor of BdE and President of SEC (CNMV) sent in advance
At least 5 attempts to locate each sample member in person
At least once on weekends.
Webpage and BdE telephone provided to interviewees to reassure on legitimacy
of survey and answer questions
9
3. FIELDWORK: Control
1. After each interview, company phoned back the respondent
Was the interview conducted? Selected responses checked following predefined script
Household composition, housing tenure, compliance with protocols and duration
BdE immediately informed about any inconsistencies
2. Recording durations (11% of the questions)
Neither interviewer nor interviewee knew which ones (except for one on reading ability)
3. Audio recording (5% of the questions)
Interviewee consent needed
Neither interviewer nor interviewee knew which ones
It has methodological interest for some difficult questions
Occupation: respondents describe what they do at their jobs and interviewer assignsISCO code using an automated search tool provided by Spanish Statistical Institute
Expected growth of the value of main residence over one year: allocate 10 points to 5 predetermined bins with possible values
Duration and audio recording allow detecting interviewers with anomalous results
Interviewers in similar locations used as a benchmark (statistical control)
Recording also allows monitoring and quick feedback if errors are detected
10
3. FIELDWORK: Validation
• Web-based system to access, review and edit each (anonymised) case
Reviewers from BdE and fieldwork company could interact on each case
• Each completed interview was reviewed by the fieldwork company, that makerecommendation to the BdE through the platform:
Editing some variables, re-contacting the respondent, or completely redoing the interview
• BdE reviews those recommendations, priority cases, and decides about alledits and re-contacts
• Doubts about financial inclusion usually led to re-contacting the respondent
Examples:
Respondent does not have a bank account, but reports receiving pension income
Respondent holds a mortgage, but reports no real estate property
Other reasons were doubts about household composition or labor market status
• Time spent on each question and the audios serve as basic controls
Help in detecting deviations from protocol (Imprecise wording, non-neutral attitude, etc)
Discipline the interviewer (McGonagle et al. 2015: “Interviewee consent to be recorded lengthens interviews by 3.5%”)
• Reviewed interviews as fieldwork progresses
Crucial to correct errors early and to provide immediate feedback to interviewers
12
4. CONCLUSIONS
• This presentation has described the contents and main characteristics of the
Survey of Financial Competences in Spain
Part of a coordinated effort by OECD to measure financial competences in 30
countries
Adapted to Spanish environment and including additional relevant features
• Described the fieldwork
Interviewer training
Measures to minimize non-response
Control during fieldwork
Validation
• Control and validation processes are always important to achieve data quality
• Specially so in a Survey of Financial Competences.
Important to distinguish if an individual does not have a financial product or,
alternatively, if she has it but did not understand the wording of the question
Control and validation processes are crucial in telling apart both cases
13
A. APPENDIX: Examples on usefulness of monitoring
Time durations of a particular question (whether or not individual is informed
about household finances) classified by whether she gave consent to record:
15
Interviewer “A” makes the question much faster when not recorded (“B” does not)
Short durations, coupled with other information, led to:
1. Immediate warning to interviewer
2. In some cases, re-visiting the household to elicit info from an informed individual